Library Automation and Digital Archive
LONTAR
Fakultas Ilmu Komputer
Universitas Indonesia

Pencarian Sederhana

Find Similar Add to Favorite

Call Number SEM 304
Collection Type Indeks Artikel prosiding/Sem
Title QR code augmented reality tracking with merging on conventional marker based backpropagation neural network
Author Gia M. Agusta, Khodijah Hulliyah, Arini, Rizal Broer Bahaweres;
Publisher 2012 International conference on advanced computer science and information systems (ICACSIS 2012) Depok, December 1st and 2nd 2012 Crystal of knowledge building Universitas Indonesia
Subject
Location
Lokasi : Perpustakaan Fakultas Ilmu Komputer
Nomor Panggil ID Koleksi Status
SEM 304 TERSEDIA
Tidak ada review pada koleksi ini: 40417
QR code Augmented reality (QRAR) is an AR doesn't require pre-registration, has 107089 combination ID-encodedandean be used on the public AR applications. From some previous research, the 6DOF tracking method less accurate, requires only small computation power and unstable marker, we propose merging a conventional marker with QR code, but it will has noise on the QR code finder patter (QRFP) under perspective distortion, in order to keep detecting so we propose a backpropagation method preceded by feature extraction with low level image processing. Method that we had been proposed achieved an accurate 6DOF, runs at 35.41 fps and resulting stable markers as a conventional marker.